Chapter Four · failure evidence
What Structural Mechanics & Stress Modeling got wrong, from 52 dissertations
Computational and analytical stress modeling frequently fails due to oversimplified kinematic representations, missing physical boundary constraints, and inadequate constitutive formulations. These modeling simplifications consistently lead to inaccurate stiffness predictions, artificial stress concentrations, and poor generalizability across loading conditions. These records come from PhD theses at 18 institutions, 2021 to 2026. Each links to its thesis. They were extracted by language models reading the full text, so treat each as a lead to read, not a verdict.
Constitutive and plasticity models fail under complex cyclic or large plastic deformation histories
Constitutive formulations that assume purely elastic inclusions, omit low amplitude cyclic events, or neglect multiaxial damage criteria fail to capture realistic plastic strain and damping. These models also suffer from systematic divergence at large plastic strains and produce unrealistic rebound or delayed pore pressure development upon loading.
Tried and failed
purely elastic inclusion models in composite plasticity applied to dislocation substructure deformation modeling. Outcome: did not generalise. Reason: failed to capture plastic deformation contributions within the cell walls of dislocation substructures
Crystal Plasticity Modeling of Heterogeneous Microstructure in Alloys · Georgia Tech
Tried and failed
deformation-based fatigue damage accumulation modeling applied to structural steel connections under arbitrary loading. Outcome: did not generalise. Reason: failed to accurately predict low-cycle fatigue under arbitrary histories outside cyclically increasing amplitudes
Parametric collapse evaluation of steel moment resisting frames with fuse connections · UT Austin
Lost to a baseline
BEM elasto-plastic analysis showed systematic divergence/loss of accuracy compared to FEM and analytical solutions at high plastic deformation across thick cylinder, notched tensile, foundation strip, and toroidal case studies.
Considered and rejected
Considered and rejected: Rejected isotropic stiffness degradation under complex stress states without proper multi-axial fracture criteria handling.
Computational investigation of fracture in polymer composites and heterogeneous dry adhesives · Imperial
Tried and failed
diffusion-driven viscoplastic constitutive modelling applied to spatially resolved dislocation density prediction. Reason: model underestimated geometrically necessary dislocation density in outer deformation zones compared to experimental measurements
Investigation of microstructural evolution of aluminium alloy 7050 at hot forging conditions · Imperial
Tried and failed
omitting low-amplitude cycles after large-amplitude loading applied to cyclic soil-structure deformation modeling. Outcome: did not generalise. Reason: small cycles caused stress redistribution and accelerated subsequent accumulation rates instead of having negligible impact
A new Cyclic Overlay Model for the Design of Laterally Loaded Monopiles · Leibniz Universität Hannover Repository
Tried and failed
calibrating cyclic nonlinear models solely on stiffness curves applied to dynamic soil-structure interaction modeling. Outcome: did not generalise. Reason: optimizing for stiffness degradation caused small-strain damping underestimation and large-strain damping overestimation
Numerical and field investigation of dynamic soil-structure interaction at full-scale · Imperial
Tried and failed
bounding surface plasticity with shear history threshold applied to soil liquefaction constitutive modeling. Outcome: worse than baseline. Reason: overly increased plastic stiffness and delayed excess pore pressure development in loose samples under monotonic loading
Seismic response of monopile foundations for offshore wind turbines in liquefiable deposits · Imperial
Considered and rejected
Considered and rejected: Rejected elastic soil models for compaction simulation because they create unrealistic deformation rebounds and internal forces upon load removal.
Soil Structure Interaction of High-Speed Trains Over Soil Metal Bridges · DalSpace
Empirical calibration and simplified test configurations fail to generalize across structural systems
Calibrating models with component scale tests, empirical reduction factors, or isolated substructure boundaries fails when applied to continuous or composite structural configurations. Furthermore, standard test configurations such as splitting or short beam shear tests break down when ductile failure and non-delamination mechanisms dominate.
Tried and failed
direct component-level parameter measurement for macro-scale calibration applied to structural dynamic building models. Outcome: did not generalise. Reason: small-scale test stiffness systematically overestimates global stiffness and underestimates fundamental oscillation periods
Tried and failed
rotational spring compliance modeling of crack depth applied to dynamic response of damaged beam structures. Reason: stiffness leveled off at deep cracks instead of vanishing at full fracture, causing large frequency errors
Tried and failed
applying empirical stiffness reduction factors across configurations applied to composite structural element stiffness modeling. Outcome: did not generalise. Reason: empirical factors calibrated for non-composite construction conditions failed under composite action with wide girder spacing
Cross-frame stiffness modification factors for composite steel I-girder bridges · UT Austin
Considered and rejected
Considered and rejected: Rejected splitting tests for strain-ductile UHPFRC tensile characterization due to improper load transfer mechanics.
Considered and rejected
Considered and rejected: Rejected ILSS short beam flexural testing for GF-PP composites because tough, ductile specimens exhibited inelastic deformation without delamination failure.
Development of high-performance thermoplastic composite laminates: : The role of interfacial adhesion · DSpace-CRIS at TU Wien
Tried and failed
calibrating linear material stiffness and boundary adjustments applied to finite element model validation. Outcome: data insufficient. Reason: field measurements and loading sequence notes had too much inherent uncertainty
Impact of lean-on bracing layouts on system stiffness behavior · UT Austin
Tried and failed
linear regression modeling of material property scaling applied to additive manufacturing infill flexural stiffness. Outcome: no signal. Reason: linear relationship between infill density and flexural modulus broke down after correcting second moment of area
Application of Co-Design Principles for Design of Series Elastic Joints · Virginia Tech
Tried and failed
isolated substructure testing under boundary approximations applied to continuous slab-column structural response. Outcome: did not generalise. Reason: isolated boundary conditions failed to capture realistic system stiffness and continuous load-rotation behavior
Lost to a baseline
Framework model with variable spring stiffness yielded higher discrepancy for dowel forces against ABAQUS (78.6 kN vs 98.5 kN) than constant stiffness model (90.0 kN vs 98.5 kN) due to ABAQUS CDP damage treatment.
The load-bearing behaviour of the steel-concrete-steel composite (SCSC) plate · DSpace-CRIS at TU Wien
Reduced dimensional element formulations fail to capture coupled 3D and out of plane deformations
Beam, frame, and 2D plane stress or strain formulations omit out of plane deformations, joint mechanics, and cross sectional ovalling. Neglecting these multidimensional kinematics leads to artificial stiffness overprediction and an inability to represent localized shear stresses or torsional degradation.
Lost to a baseline
Equivalent Frame Analysis (EFA) required an artificially high modulus of elasticity (E=6450 MPa vs E=1800 MPa for shell elements) and failed to capture out-of-plane wall deformations.
Performance Assessment and Retrofit Strategies for Unreinforced Masonry Structures · YorkSpace
Considered and rejected
Considered and rejected: Fiber-element formulation (used in SAFIR/OpenSees) rejected because it only accounts for axial normal stresses and cannot properly capture shear stresses and torsional stiffness degradation in thin-walled open sections.
Development and application of corotational finite elements for the analysis of steel structures in fire · IRIS - UNITN - prod
Considered and rejected
Considered and rejected: Rejected standard 2D shell elements in FEM simulations because they fail to capture the realistic nearly-incompressible hyperelastic and visco-elastic deformation modes of rubbery photopolymer creases.
Design of Kresling inspired architected materials by using Polyjet technique · IRIS - UNITN - prod
Tried and failed
beam finite element modelling applied to dense lattice microstructures. Reason: neglects joint intersection mechanics at high volume fractions, causing severe stiffness over-prediction
Multiscale structural optimisation with concurrent coupling between scales · Imperial
Tried and failed
corotational beam formulation applied to inflatable low-modulus thin-walled beams. Reason: overpredicts stiffness by ignoring membrane wrinkling and cross-sectional ovalling effects
Analysis, Design, and Experimentation of Beam-Like Structures · Virginia Tech
Considered and rejected
Considered and rejected: Rejected Euler-Bernoulli and Timoshenko beam theories due to inability to capture 3D coupled translational-rotational modes (axial/torsional) and large finite deformations without artificial elongation
Geometric Recursive Dynamic Modelling and Simulation of Soft Robotic Systems · Carleton University Institutional Repository
Considered and rejected
Considered and rejected: Rejected membrane-only analysis models due to significant bending and shear deformation contributions in thick and moderately thin plate specimens
Experimental Investigation Of Polymer Adhesion Mechanics Using A Blister Contact Test · Penn
Tried and failed
2D plane stress and strain finite element modelling applied to mechanical rolling deformation and stress. Reason: failed to capture 3D out-of-plane lateral deformation, mispredicting residual stress and plastic strain distributions
Analytical and homogenized micromechanical models overestimate structural stiffness and strength
Analytical formulations and simplified representative volume elements overestimate stiffness and load carrying capacity by misrepresenting internal load transfer and microstructural size effects. Perturbation assumptions and neglect of boundary free surfaces create large deviations from experimental measurements and detailed numerical baselines.
Lost to a baseline
GLS analytical model overestimated composite tensile failure strength by about 30% relative to experimental measurements
Process development of pre-impregnated hybrid thin-ply composites · EPFL
Lost to a baseline
Lumped Element Model (LEM) overestimates effective stiffness and resonant frequency compared to the Stiffness Matrix Method (SMM) due to perturbation approximations of mass loading.
STL NEMS fabrication, design, and inspection · Imperial
Considered and rejected
Considered and rejected: Long RVE models where load is applied directly to inclusions without matrix-mediated interphase transfer, rejected because they fail to represent realistic load transfer and overestimate stiffness.
Study on the Mechanical Properties of Carbon Nanotube Coated‒Fiber Multi-Scale (CCFM) Hybrid Composites · IRIS - POLITO - prod
Tried and failed
classical higher-order shear deformation plate theory applied to microstructured metamaterial dynamic characterisation. Reason: omitting non-classical length scale parameters ignores microstructural size-effects, underpredicting natural frequencies
Static and dynamic characterisation of the Zero Poisson's radio fish cells metamaterial PhD in Aerospace · Cranfield
Considered and rejected
Considered and rejected: Rejected pill-shaped cross-section model for beam bending moment of inertia calculations because its resulting Young's modulus values significantly diverged from uniaxial tensile test results.
Lost to a baseline
Experimental elastic moduli underperformed classical laminate plate theory (CLPT) apparent stiffness predictions across all layups (e.g., 24.5% difference for consolidated UD-90°)
Lost to a baseline
Peri-CORE local stress predictions deviated by 15-20% from FE results near free surfaces in uniaxial tensile loading.
Peridynamics and finite element simulations of TRISO Fuel under extreme operating conditions · Imperial
Lost to a baseline
CPT-based monotonic tensile capacity predictions significantly overestimated field measured capacities for small-diameter driven piles at Allen Harbor.
An investigation of factors affecting interface shear resistance between pipes and soils · UT Austin
Omitting initial preloads root fillets or realistic boundary constraints induces numerical distortion
Neglecting physiological pre-strain or static preload contact forces distorts structural deformations and corrupts reduced dynamic stiffness matrices. Additionally, omitting geometric fillets or applying overly rigid boundary constraints generates artificial stress singularities and incorrect stiffness responses.
Tried and failed
finite element modeling without physiological pre-strain applied to heart valve mechanical closure simulations. Reason: omitting initial pre-strain caused non-physiological deformation and failed to match in vivo strain distributions
Investigate mitral valve regurgitation following myocardial infarction using cardiac in-silico models · UT Austin
Tried and failed
linear static finite element analysis applied to truss structure deflection modeling. Outcome: worse than baseline. Reason: linear static models exhibited artificial stiffness leading to underestimated deflections under loading
Active Force Correction of Off-Nominal Structures Using Intelligent Scaffolding · Virginia Tech
Tried and failed
Finite element analysis with rigid boundary constraints applied to mechanical joint stress modeling. Reason: Stress singularities at fixed boundary conditions artificially exceeded the material ultimate tensile strength
Tried and failed
omitting fillet geometry in finite element models applied to cross-section torsional stiffness estimation. Outcome: worse than baseline. Reason: neglecting root fillets led to significant discrepancies between numerical and analytical torsional stiffness
Stability and design of steel angle section members · Imperial
Tried and failed
omitting secondary structural constraints from objective function applied to biomechanical PDE-constrained inversion. Outcome: worse than baseline. Reason: loss of displacement regularization severely degraded anatomical deformation accuracy
High-performance inversion algorithms for brain tumor growth models in personalized medicine · UT Austin
Tried and failed
component mode synthesis without preloaded contact forces applied to constrained non-linear structural dynamics. Outcome: worse than baseline. Reason: omitting static preload contact forces creates an inaccurate reduced stiffness matrix and inaccurate displacement predictions
Structural Dynamics Design of Steam Turbine Blades with Friction Contacts for Renewable Energy · IRIS - POLITO - prod
Inappropriate discretization order or excessive solver complexity degrades computational performance
Static and first order finite element analyses underpredict time varying displacement peaks and yield elevated strain errors compared to higher order models. Conversely, resolving full 3D domains or continuous damage metrics adds heavy computational overhead without accuracy improvements over simpler baselines.
Tried and failed
static finite element analysis applied to time-varying vessel wall displacement. Outcome: worse than baseline. Reason: failed to capture time-varying evolution of deformation, under-predicting peak displacement
Lost to a baseline
FDM produced an accurate result faster than FEM for shell analyses governed by membrane stress fields or smooth deformation gradients (BOSOR3 comparison by Bushnell).
Modern analysis of axisymmetric shells · Imperial
Lost to a baseline
Instantaneous stiffness degradation (USDFLD) progressive failure approach is ~8 times faster than the continuous energy-based Matzenmiller UMAT approach with Hashin criteria for braided SVE tensile modeling while yielding negligible discrepancy.
Metamodel-based uncertainty quantification for the mechanical behavior of braided composites · Leibniz Universität Hannover Repository
Considered and rejected
Considered and rejected: Rejected solving deformation gradient metrics via semidefinite programming per time-step due to computational cost, replacing it with an analytic optimal metric formula.
Robustness analysis of continuous-depth neural networks · DSpace-CRIS at TU Wien
Considered and rejected
Considered and rejected: Rejected full 3D finite element simulation for the parametric study in favor of 2D axisymmetric modeling after verifying identical stress/deformation responses with lower computational cost.
Effectiveness of geogrids in roadway construction; determine a granular equivalent (G.E.) factor · Iowa State
Lost to a baseline
1st order FEA strain predictions had errors up to 12.6% compared to 2nd order FEA which reduced errors to within 7.2% across single-angle tensile and compressive tests.
Cross-frame stiffness modification factors for composite steel I-girder bridges · UT Austin
Left open by the authors
Problems the authors named and did not get to.
Left open
Develop a 3D finite element model simulating bacterial cell deformation inside non-axisymmetric tapered microfluidic channels with corners. Blocker: Requires empirical cross-sectional geometry data and experimental deformation images from microfluidic extrusion experiments
DECIPHERING BACTERIAL CELL ENVELOPE MECHANICS AND MECHANOSENSING: A MECHANICAL ENGINEERING APPROACH · Cornell
Left open
Derive bending stiffness reduction factors across varied tunnel lining geometries, joint configurations, and depths using finite element modeling. Blocker: Requires the proprietary Imperial College Finite Element Program (ICFEP) codebase and the specific custom constitutive model implementations.
Development of advanced numerical models for grey cast iron tunnel linings · Imperial
Left open
Simulate droplet dimensionality and actin bending rigidity parameters in models of actin ring-to-rod droplet deformation transitions. Blocker: Lacks specific mathematical models, simulation targets, or precise parameter ranges to implement directly
Liquid-like condensates determine actin network organization · UT Austin
Left open
Investigate buckling load-displacement characteristics within the crash deformation framework. Blocker: Lacks specification of the exact crash deformation framework and finite element model parameters
High-Dimensional Generative Models for 3D Perception · Virginia Tech
Left open
Perform mesh convergence analysis over perforated hole regions and apply boundary constraints to limit unphysical hyperelastic tissue deformation in Abaqus. Blocker: Requires the proprietary patient-specific CAD geometry and specific Abaqus model setup developed in the thesis.
Design and Manufacturing of a Novel Intraoral Prosthesis for Obstructive Sleep Apnea · MIT
Left open
Integrate stress analysis into the multi-objective genetic algorithm to constrain maximum tensile and shear stresses during TMD optimization. Blocker: Requires access to proprietary simulation software (MSC NASTRAN/MATLAB) and specific FEM model files from the thesis.
Dynamic Aeroelastic Response Attenuation of Aerospace Structures Employing Passive Tuned Mass Dampers · Carleton University Institutional Repository
Left open
Extend and validate the Cyclic Strain Accumulation Method for modeling cyclic foundation deformation in cohesive soils. Blocker: None
A new numerical concept for the prediction of cyclic deformation accumulation of foundation structure in non-cohesive soil · Leibniz Universität Hannover Repository
Checking a claim in this area?
We can run the same search on any method or claim. If nothing turns up, we will say so, and that proves nothing on its own.